1. Field of the Invention
The present invention relates to a solid-state image pickup device, and particularly to an MOS type solid-state image pickup device and a method of driving the MOS type solid-state image pickup device.
2. Description of the Related Art
MOS type solid-state image pickup devices are known as one type of solid-state image pickup devices. Further, as one of the MOS type solid-state image pickup device is known a solid-state image pickup device of transferring signal charges of photodiodes constituting pixels to a detection portion using transfer transistors and then reading out the signal charges. This type of MOS type solid-state image pickup device has a CMOS logic circuit in the same chip and the pixels are operated with the same single low-voltage power source as the logic circuit unlike the CCD type solid-state image pickup device. Accordingly, for example when the transfer transistors are n-channel MOS transistors, the gate voltage of the transfer transistor in each pixel has two values of 0V and a power resource voltage Vdd.
Two components, one of which is signal charge corresponding to the amount of incident light and the other of which is a dark current component (dark electrons) flowing in the photodiode of each pixel even when no light is incident, are accumulated in the photodiodes during a charge accumulation period. The dark electrons are not separated from the signal charge in the reading operation, and dispersion of the dark electrons causes noises. Particularly, the dispersion of dark current every pixel causes a fixed pattern noise, and an image is achieved as if it is picked up through frosted glass. Further, the time-dispersion of the dark current causes a random noise. Accordingly, it is an important problem for the MOS type solid-state image pickup device to reduce the dark current at maximum.
An embedded photodiode is known as a conventional technique of reducing the dark current. The most dark current generating source is the interface between Si and SiO2 above the photodiode. If this area is depleted, dark current would flow into the photodiode. Accordingly, the oxide film interface on an n-type semiconductor region constituting the photodiode is neutralized by a p-type region to suppress occurrence of dark current.
FIG. 11 shows a part of a pixel having a conventional embedded photodiode, that is, shows the cross-sectional structure of an embedded photodiode and a transfer transistor.
As shown in FIG. 11 a p-type semiconductor well region 72 is formed on an n-type or p-type semiconductor substrate 71, and a photodiode PD and a transfer transistor QT are formed in a pixel area sectioned by an element separating layer (SiO2 layer) 73 formed by selective oxidation (LOCOS) of the p-type semiconductor well region 72, thereby forming a pixel 70, that is, the main part thereof. The photodiode PD is formed as a so-called embedded photodiode by forming an n-type semiconductor region 74 serving as a charge accumulating area in the p-type semiconductor well region 72 and then forming a p-type semiconductor region 75 having the opposite conduction type to the n-type semiconductor region 74 on the upper surface of the n-type semiconductor region 74. The transfer transistor QT is constructed by setting the n-type semiconductor region 74 of the photodiode PD as one source/drain region and forming a transfer gate electrode 78 through a gate insulating film (for example, SiO2 film) 77 between the region 74 and the other n-type source/drain region 76 formed in the p-type semiconductor well region 72. The other n-type semiconductor region 76 of the transfer transistor QT is constructed as a detection portion.
In this construction, the p-type semiconductor region 75 is formed at the interface between the n-type semiconductor region 74 and the oxide film 79 above the n-type semiconductor region 74 constituting the photodiode PD, so that the depletion of the interface can be prevented and occurrence of dark current from the interface can be suppressed. This photodiode is the embedded photodiode.
The MOS type solid-state image pickup device is required to operate with a low voltage. For example, as compared with the power source voltage of 12V for the CCD type solid-state image pickup device, a low power source voltage of 3V or less is needed to the MOS type solid-state image pickup device. In this case, signals of photodiodes are read out only within such a low voltage range that the signals can be transmitted from the photodiodes, so that it is difficult to keep a sufficient saturation signal amount. Accordingly, the dynamic range is small and the gradation cannot be adjusted.
On the other hand, occurrence of dark current from the oxide film interface of the photodiodes is suppressed by the embedded photodiodes, however, the other residual dark current components produce fixed pattern noises and random noises under low illumination. Therefore, more improvements have been required to be made for MOS type solid state image pickup devices in order to have high S/N and high sensitivity.
Further, in the MOS type solid-state image pickup device, when light having a light amount above saturation level is incident to the photodiodes PD, photoelectrically-converted signal charges, that is, photoelectrons in this case overflow from the photodiodes PD. The photoelectrons thus overflowing are diffused and expanded in the p-type semiconductor well region 72, and invade into the photodiodes PD of the surrounding pixels to cause false signals. This phenomenon is called as “blooming”. In order to prevent the blooming, the overflow path of the overflowing photoelectrons (so-called as “overflow path”) is placed in advance to prevent diffusion of the photoelectrons into the p-type semiconductor well region 72.
In the prior art, the overflow path is set in the channel portion of the transfer transistor QT, and the photoelectrons overflowing from the photodiodes PD are made to flow out through the channel portion of the transfer transistor QT into the detection portion (n-type source/drain region 76) side. The overflowing photoelectrons are made to flow out to a reset transistor side by the detection portion as not shown.
However, in this construction, as the potential of the channel portion of the transfer transistor QT is lowered, that is, when 0V is applied to the transfer gate electrode 78, the potential of the channel portion is equal to about 0.5 to 0.6V, and as the potential of this channel portion is lowered to 0V side, the amount of charges which can be passed through the overflow path is reduced.
In the MOS type solid-state image pickup device, the dynamic range is enlarged by increasing the saturation signal amount. Therefore, it is required that the amplitude of the gate voltage to be applied to the transfer gate is set to a large value, that is, the potential of the channel portion of the transfer transistor during the charge accumulation period is lowered. However, the conventional MOS type solid-state image pickup device has a problem that the enlargement of the dynamic range by increasing the saturation signal amount is incompatible with the ensuring of the performance of the overflow path as described above.
The foregoing description is made on the assumption that electrons are set as signal charges and the n-channel MOS transistor is used as the transfer transistor. However, the same problem also occurs on the assumption that holes are set as signal charges and a p-channel MOS transistor having the opposite conduction type is used as the transfer transistor.